Senior Applied Scientist, Amazon Core Search

Role Overview

Amazon Search is on a multi-year mission to enhance the global shopping journey by delivering intuitive and efficient search experiences. Our team focuses on designing customer-centric search technologies that help users find products seamlessly across Amazon's extensive catalog, supporting multiple languages and use cases from the very first keystroke.

As the scope of Amazon Search expands to new interfaces and devices, maintaining excellence in Search Assistance and Search Quality is crucial. We are seeking a Senior Applied Scientist to pioneer improvements in search autocomplete using advanced NLP, Machine Learning (ML), and Deep Learning (DL) techniques.

Primary Responsibilities

• Lead innovations in search autocomplete by creating high-quality, diverse suggestion models to enhance customer experience.
• Develop and train semantic matching models including bi-encoders, cross-encoders, and methods leveraging knowledge distillation from large foundational models for effective suggestion ranking.
• Implement reinforcement learning and reward-based modeling techniques for ongoing enhancement of autocomplete relevance and quality.
• Design multi-objective ranking and scoring frameworks to balance aspects like diversity, specificity, and relevance of search suggestions.
• Architect scalable ML models optimized for stringent latency requirements, employing strategies such as knowledge distillation, quantization, and efficient inference suitable for production environments.
• Lead full lifecycle science projects, from problem definition to deployment, while mentoring colleagues and collaborating cross-functionally to ensure impactful customer outcomes.

Required Qualifications

• PhD or Master’s degree in a relevant field.
• Over 6 years of hands-on experience in building machine learning models or algorithm development for business applications.
• Proficiency in programming languages such as Java, C++, Python, or equivalent.
• Strong background in neural deep learning methodologies and machine learning techniques.

Preferred Skills

• Familiarity with ML frameworks and tools including R, scikit-learn, Spark MLLib, MxNet, TensorFlow, numpy, scipy, and the like.
• Experience working with large-scale distributed processing systems such as Hadoop and Spark.

Additional Information

Amazon values inclusivity and strives to create an accommodating workplace culture. Applicants with disabilities are encouraged to request accommodations during the recruitment process to support interview and onboarding needs.